Google's new Flash TTS models let you design AI voices from scratch using text descriptions

AI Digest

Google推出Flash TTS和Flash-Lite TTS模型，支持100+语言，用户可通过文本描述创建AI声音，兼具语音克隆和场景化对话功能，即将通过API开放。

Google's Flash TTS and Flash-Lite TTS models enable voice creation from text descriptions across 100+ languages, with voice cloning and scene-directed dialogue, set to launch via APIs.

Key points

• 支持100+语言的文本生成语音，可定制音色、口音等特征 Text-to-speech models support over 100 languages with customizable voice traits
• 提供语音克隆功能，30秒音频可生成个性化声纹 Voice cloning feature creates profiles from 30-second audio samples
• 支持场景化对话控制，包含笑声、叹气等非语言声音 Scene-directed dialogue controls include laughter and sighs
• 通过Gemini API和AI Studio逐步开放，企业版API即将推出 Models roll out via Gemini API with enterprise access pending
• 按需计费模式，音频输出成本约0.54-1.62美元/小时 Pricing based on text/audio tokens with cost estimates provided

Takeaway: AI语音生成进入高度定制化时代，文本描述即可创建独特声线。 / AI voice generation reaches new customization levels through text-based voice creation.

Why it matters 揭示AI语音生成技术突破，为开发者提供新工具和商业机会。